The ability to transform plants by integrating and expressing desirable polynucleotides in plant cells makes it possible to efficiently introduce agronomic and quality traits into a variety of plant species. Transgenic plants that are produced by current transformation methods, however, require extensive tissue culture manipulations, which are time consuming and species specific. Furthermore, such methods do not only integrate the desirable polynucleotide(s) into a plant's genome, but also additional and superfluous nucleic acids. When making a genetically engineered food, the superfluous nucleic acids may be undesirable because they are from non-food sources, such as viruses and bacteria and are, therefore, undesirable.
Existing plant transformation methods rely on the use of Agrobacterium for DNA transfer. These methods typically comprise (1) preparing tissue explants, (2) infecting explants with at least one disarmed Agrobacterium strain, (3) culturing and selecting the transformed plant cells on tissue culture media, and (4) inducing proliferation and subsequent regeneration to generate whole plants. Examples of these methods are described in U.S. Pat. Nos. 5,591,616, 6,051,757, 5,164,310, and 5,693,512, and EP 0 672 752 A1, which are incorporated herein by reference. However, explant preparation is a laborious process that requires extensive resources, especially for many monocotyledonous plant species including maize, wheat, barley, and oats.
Furthermore, the subsequent process of proliferation and regeneration is also very laborious, taking at least 12 months to develop a primary transformed plant. Since different plants require different concentrations of salts, minerals, and hormones, including auxins and cytokinins, for proliferation and regeneration, the applicability of typical transformation methods is limited to one species or only a few cultivars of one species.
Even by optimizing cultivar-specific transformation methods, successful transformation has been accomplished for only a very few cultivars of important crop species, such as for the maize inbred lines H99, Oh43, and B73, the spring wheat variety Bobwhite, and the cotton cultivar Coker 312. The introduction of foreign DNA into elite germplasm often requires the transformation of inferior cultivars followed by conventional multi-year breeding programs to introgress the DNA into the desired material.
Tissue culture manipulations can be avoided by either vacuum infiltrating plants with an Agrobacterium suspension or emerging such plants in suspensions that also contain approximately 0.05% Silwet L-77 (Bechtold et al., Acad Sci Paris Life Sci 316: 1194-1199, 1993; Clough & Bent, Plant J 16: 735-743, 1998). However, this method is only applicable to the model plant systems Arabidopsis thaliana, Arabidopsis lasiocarpa, and Raphanus sativus. Transgenic plants can also be obtained for a fourth plant species, Medicago trunculata, by vacuum infiltrating seedling with Agrobacterium suspensions.
Such in planta transformation systems are of limited utility, however, and not applicable to commercially relevant crop plants. Efforts to broaden such applicability to encompass a larger variety of crops have failed because of the inaccessibility of those crops to Agrobacterium-mediated transformation, and/or the resultant, detrimental physiological responses, such as flower abscission and Agrobacterium-induced necrosis.
Alternative transformation systems include direct DNA delivery systems like particle bombardment (U.S. Pat. No. 4,945,050), polyethylene glycol treatment (U.S. Pat. No. 6,143,949), microinjection (U.S. Pat. No. 4,743,548), whiskers (U.S. Pat. No. 5,302,523), and electroporation (U.S. Pat. No. 5,284,253). Whereas DNA transfer mediated by Agrobacterium is often limited to one to three copies of foreign DNA, direct DNA delivery systems usually result in the transfer of many more copies, which may integrate randomly throughout the plant genome. The unnecessary abundance of insertions is undesirable and may negatively affect the plant genome's integrity.
Sonication was shown to greatly enhance the efficiency of both Agrobacterium-mediated transformation and direct DNA delivery (U.S. Pat. No. 5,693,512). The ultrasound vibrations are believed to disrupt cell walls and thereby facilitate foreign DNA transfer. Sonication reduces the viability of tissue explants, and any increase in transformation frequency may be compromised by an increase in non-viable or dying plants.
These, as well as more conventional transformation methods, introduce a variety of viral and bacterial genetic elements into plant cells. At least four different genetic elements, derived from bacteria, are typically used to transform plants (During, Transgenic Research 3: 138-40, 1994). Such elements include regulatory sequences such as promoters and terminators to promote appropriate transgene expression in plants. An example of a frequently used foreign promoter is the 35S “super” promoter of Cauliflower Mosaic Virus (CaMV), which is able to not only induce high levels of expression of the transgenes but also enhance the expression of native genes in its vicinity (Weigel et al., Plant Physiol., 122: 1003-13, 2000).
Other strong viral promoters include those from rice tungro bacilliform virus, maize streak virus, cassava vein virus, mirabilis virus, peanut chlorotic streak caulimovirus, figwort mosaic virus and chlorella virus. Other frequently used promoters are derived from bacterial species and include the promoters of the nopaline synthase and octopine synthase gene. Only a few strong and constitutive promoters are derived from food sources. Examples of such promoters are the promoters of the maize Ubiquitin-1 gene (U.S. Pat. No. 6,054,574; and WO 01/94394), the sugarcane Ubiquitin-4 gene (U.S. patent application Ser. No. 02/0046415), and the potato Ubiquitin-7 gene (Garbarino et al., U.S. Pat. No. 6,448,391 B1, 2002). The applicability of most other plant promoters is limited because of low activity, tissue specificity, and/or poor developmental regulation. Typical terminators are those associated with the nopaline synthase and octopine synthase genes from Agrobacterium. 
Also required for transformation is the Agrobacterium-derived transfer DNA, i.e., the T-DNA, which transfers desired polynucleotide(s) from Agrobacterium into plant cell genomes. Thus, transgenic plants of the conventional art contain much superfluous foreign DNA. Furthermore, the infidelity of DNA transfer can result in co-integration of bacterial plasmid sequences that are adjacent to the T-DNA. In fact, about 75% of transformation events in plants such as tomato, tobacco, and potato may contain such superfluous plasmid backbone DNA (Kononov et al., Plant J. 11: 945-57, 1997). The presence of backbone sequences is undesirable because they contain bacterial origins of replication and/or encode for antibiotic resistance genes.
Thus, there is a need for accelerated and species-independent methods for transferring and expressing desired polynucleotides into plant cells and genomes. There is also a need to limit the co-transfer of superfluous, undesirable DNA, if the target plant is a food crop. Such methods are provided herein. To optimize DNA transfer from Agrobacterium to individual plant cell nuclei, plant tissues such as seedlings are agitated in an Agrobacterium suspension. To optimize the subsequent integration of the transferred DNAs into the genome of plant cell nuclei, the plant tissues are exposed to chemicals that induce double strand breaks. Vectors are used that are designed to limit the transfer of undesirable DNA.